[PATCH v5 29/36] KVM: arm64: Use page-table to track page ownership

[Quentin Perret]

On Monday 15 Mar 2021 at 16:36:19 (+0000), Will Deacon wrote:
> On Mon, Mar 15, 2021 at 02:35:29PM +0000, Quentin Perret wrote:
> > As the host stage 2 will be identity mapped, all the .hyp memory regions
> > and/or memory pages donated to protected guestis will have to marked
> > invalid in the host stage 2 page-table. At the same time, the hypervisor
> > will need a way to track the ownership of each physical page to ensure
> > memory sharing or donation between entities (host, guests, hypervisor) is
> > legal.
> > 
> > In order to enable this tracking at EL2, let's use the host stage 2
> > page-table itself. The idea is to use the top bits of invalid mappings
> > to store the unique identifier of the page owner. The page-table owner
> > (the host) gets identifier 0 such that, at boot time, it owns the entire
> > IPA space as the pgd starts zeroed.
> > 
> > Provide kvm_pgtable_stage2_set_owner() which allows to modify the
> > ownership of pages in the host stage 2. It re-uses most of the map()
> > logic, but ends up creating invalid mappings instead. This impacts
> > how we do refcount as we now need to count invalid mappings when they
> > are used for ownership tracking.
> > 
> > Signed-off-by: Quentin Perret <qperret at google.com>
> > ---
> >  arch/arm64/include/asm/kvm_pgtable.h |  21 +++++
> >  arch/arm64/kvm/hyp/pgtable.c         | 127 ++++++++++++++++++++++-----
> >  2 files changed, 124 insertions(+), 24 deletions(-)
> > 
> > diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h
> > index 4ae19247837b..683e96abdc24 100644
> > --- a/arch/arm64/include/asm/kvm_pgtable.h
> > +++ b/arch/arm64/include/asm/kvm_pgtable.h
> > @@ -238,6 +238,27 @@ int kvm_pgtable_stage2_map(struct kvm_pgtable *pgt, u64 addr, u64 size,
> >  			   u64 phys, enum kvm_pgtable_prot prot,
> >  			   void *mc);
> >  
> > +/**
> > + * kvm_pgtable_stage2_set_owner() - Annotate invalid mappings with metadata
> > + *				    encoding the ownership of a page in the
> > + *				    IPA space.
> 
> The function does more than this, though, as it will also go ahead and unmap
> existing valid mappings which I think should be mentioned here, no?

Right, I see why you mean. How about:

'Unmap and annotate pages in the IPA space to track ownership'

> > +int kvm_pgtable_stage2_set_owner(struct kvm_pgtable *pgt, u64 addr, u64 size,
> > +				 void *mc, u8 owner_id)
> > +{
> > +	int ret;
> > +	struct stage2_map_data map_data = {
> > +		.phys		= KVM_PHYS_INVALID,
> > +		.mmu		= pgt->mmu,
> > +		.memcache	= mc,
> > +		.mm_ops		= pgt->mm_ops,
> > +		.owner_id	= owner_id,
> > +	};
> > +	struct kvm_pgtable_walker walker = {
> > +		.cb		= stage2_map_walker,
> > +		.flags		= KVM_PGTABLE_WALK_TABLE_PRE |
> > +				  KVM_PGTABLE_WALK_LEAF |
> > +				  KVM_PGTABLE_WALK_TABLE_POST,
> > +		.arg		= &map_data,
> > +	};
> > +
> > +	if (owner_id > KVM_MAX_OWNER_ID)
> > +		return -EINVAL;
> > +
> > +	ret = kvm_pgtable_walk(pgt, addr, size, &walker);
> > +	dsb(ishst);
> 
> Why is the DSB needed here? afaict, we only ever unmap a valid entry (which
> will have a DSB as part of the TLBI sequence) or we update the owner for an
> existing invalid entry, in which case the walker doesn't care.

Indeed, that is now unnecessary. I'll remove it.

Thanks,
Quentin